Streptococcus pneumoniae is an important pathogen, causing invasive diseases such as pneumonia, meningitis and bacteraemia. Even in regions where effective antibiotic therapy is freely available, the mortality rate from pneumococcal pneumonia can be as high as 19% in hospitalised patients. In developing countries, in excess of 3 million children under the age of 5 years die each year from pneumonia, of which S. pneumoniae is the commonest causative agent. S. pneumoniae also causes less serious, but highly prevalent infections such as otitis media and sinusitis, which have a significant impact on healthcare costs in developed countries. Otitis media is especially important in young children, while sinusitis affects both children and adults.
The heptavalent polysaccharide conjugate vaccine from Wyeth, sold as Prevnar® in the United States and as Prevenar® in the rest of the world, is currently the only effective conjugate vaccine available for protection against Streptococcus pneumoniae infection (Kyaw et al, 2002; Hausdorff et al, 2000). The vaccine comprises seven purified Streptococcus capsular polysaccharides (serotypes 4, 6B, 9V, 14, 18C, 19F and 23F) out of a possible 90 (Kalin, 1998), each conjugated to a carrier protein. Preparation of such a vaccine is described in U.S. Pat. No. 4,673,574 (Anderson). The protein used for conjugation of the capsular polysaccharides is a diphtheria toxoid, CRM197, offering an increase in the immunogenicity of the vaccine in infants (Blum et al, 2000; Katkocin, 2000). However, each serotype of S. pneumoniae has a structurally distinct capsular polysaccharide, such that immunization with one serotype tends not to confer protection against the majority of the other serotypes, although some cross-protection does occur against vaccine-related serotypes (Whitney et al., 2003).
Complementary approaches to serotype-specific immunization are being investigated. A possibility is to also use a species-common virulence factor such as Pneumolysin (PLY), the 53 kDa toxin produced by all invasive strains of S. pneumoniae (Paton et al, 1993). PLY could be used alone or as a carrier protein conjugated to the polysaccharides in Prevnar®, offering increased efficacy. Alexander et al (1994) demonstrated that immunisation of mice with a PLY toxoid conferred immune protection upon Challenge with 9 different serotypes of S. pneumoniae. PLY has been shown to stimulate an immune response similar to that of S. pneumoniae infection by activating the classical complement pathway (Paton et al, 1984) and inducing apoptosis of neutrophils and macrophages (Cockeran et al, 2002; Kadioglu et al, 2000).
PLY belongs to the group of Cholesterol-binding Cytolysins (CBCs) that bind to the cholesterol of host cell membranes prior to formation of large 30-50mer ring structures that create lytic pores (Palmer, 2001; Jedrzejas, 2001). The mechanism of pore-formation is not fully understood and there is much debate over the sequence of events (Bonev et al, 2000; Shepard et al, 1998). However, the ability to form pores means that native PLY is highly toxic, which is a problem in terms of the development of immunogenic compositions.
Although the conjugation process used in production would render PLY non-toxic, it would be more favourable to start with a non-toxic form. Further, a toxic form would be difficult to use in preparation of unconjugated immunogenic compositions. The toxicity of PLY can be significantly reduced by site-directed mutagenesis to create PLY toxoids, known as pneumolysoids (Paton, 1996).
A variety of such toxoids exist and have been shown to give immune protection, either independently or when conjugated to polysaccharides, to mice in response to a challenge with virulent type 2 D39 S. pneumoniae (Paton et al, 1991; Alexander et al, 1994). Most mutations have previously been created in the highly conserved 11 amino acid region near the C′ terminus (Mitchell et al, 1992; Berry et al, 1995). This site has been shown to be involved in binding to the host cell (de los Toyos et air 1996). A number of such mutated forms of PLY are described in International Patent Application WO 90/06951; each of the mutations described in this publication is towards the C′ terminus of the protein.
A further problem with PLY is that it aggregates upon large-scale production, a problem which must be solved in order for PLY to be used in immunogenic compositions. It is believed that the aggregation of PLY is related to the oligomerisation of PLY involved in pore formation. The present invention thus attempts to reduce or eliminate PLY-PLY interaction (oligomerisation), such that the chance of aggregation during large-scale production will decrease, thereby creating an easily purified form of PLY.
Toyos et al (1996) describe the raising of monoclonal antibodies (mAbs) to various regions of PLY, and probing of the whole toxin and a ‘proteinase K nicked’ form. Proteinase K cuts PLY into a 37 kDa and 15 kDa fragment. Antibody mAb PLY 4 only recognised whole PLY, and neither of the fragments, indicating that the epitope on PLY for this mAb is within the nicked region. When PLY was pre-incubated with mAb PLY 4, then added to liposomes, the toxin no longer formed pores on the liposome membrane. This implies that the site blocked by mAb PLY 4 (thought to be the Asparagine N143 region) is the site responsible for interaction with other PLY monomers to form oligomeric pores. Oligomerisation of Streptolysin O from Streptococcus sp. can also be blocked by mAbs as demonstrated by Hugo et al in 1986. It is unknown whether the antibodies directly prevent oligomerisation by binding to the oligomerisation site or if there is an association that sterically hinders the interaction of toxin monomers.
Monoclonal antibody PLY 4 has been further characterised by Suárez-Álvarez et al (2003) and they suggest that the epitope for mAb PLY 4 is further downstream than the N143 region initially proposed by Toyos et al in 1996. The site of recognition now appears to be conformation dependent and within amino acids E151-Y247 and not within the N143 region.
Previously a N142N143 deletion and N143D substitution within PLY were created by the present inventors as initial steps to understanding this region and its role in oligomerisation. Characterisation of both mutants revealed identical behaviour to native PLY in terms of haemolysis and pore formation (Search, 2002), suggesting that oligomerisation was not blocked, and the toxicity of the mutants remained unchanged. Thus, previously created mutant PLY forms do not exhibit reduced toxicity or reduced oligomerisation, suggesting that these mutations will not be of assistance in the production of immunogenic compositions.